Human retinal ganglion cell sub-populations have not been systematically studied with modern techniques. There is compelling evidence in animal species that ganglion cell shape is correlated with function. We propose to create a morphologic taxonomy of the human ganglion cells by applying new techniques that have been developed for staining ganglion cells and filling their dendrites with marker molecules. Specifically, we will utilize fluorescent dyes and single cell injections to characterize these cells. These fluorescent dyes have the ability to selectively stain sub-populations of cells, thus permitting rapid identification of cell types over wide areas of the whole-mount retina. Single-cell injections are used to definitively demonstrate the cell shape. This technique has the special advantage of being applicable to post-mortem tissue, thereby making it possible to study human retina. Clinical experience suggests that optic neuropathies represent a heterogenous group of disorders. For example, in a disease such as glaucoma, the early neuropathy may be due to a selective vulnerability of certain sub-classes of ganglion cells. If this is so, there may be (in this condition and others) sub-clinical visual pathology not detected by the usual visual tests because only a few functional types of retinal ganglion cells are damaged. As an initial approach to the question, we will follow the loss of identified ganglion cell types in rabbits after experimental injury to the optic nerve. Better characterization of human ganglion cell types is of interest in its own right, and also for comparison with the well characterized ganglion cell populations in animals. If it is true that diseases of the optic nerve may preferentially affect selected types of retinal ganglion cells, this knowledge may lead to the development of more specific tests of visual function and the detection of more subtle disease states.